BusinessFriday, May 1, 2026· 2 min read

Pentagon Clears OpenAI, Google, Nvidia and More to Use AI on Classified Networks

Source: The Verge AI

TL;DR

The Pentagon has approved major AI vendors — including OpenAI, Google, Microsoft, Amazon, Nvidia, xAI and Reflection — to operate on classified networks, expanding secure government access to advanced AI tools. The move signals growing institutional trust in commercial AI, while supply-chain vetting led to Anthropic's exclusion for now.

Key Takeaways

  • 1Multiple leading AI companies have been authorized for use on U.S. classified networks, enabling direct government access to advanced models.
  • 2This is a practical step toward integrating commercial AI into national defense workflows, not just exploratory research.
  • 3Anthropic was excluded after a supply-chain risk designation, showing rigorous security vetting for classified use.
  • 4The agreements build on earlier deals and reflect accelerating collaboration between the defense sector and private AI innovators.
  • 5Authorized access could speed development of secure, mission-ready AI tools for analysts, operators, and decision-makers.

Major AI Vendors Approved for Classified Use

The Pentagon has finalized agreements with several leading AI providers — including OpenAI, Google, Microsoft, Amazon, Nvidia, Elon Musk's xAI, and startup Reflection — to allow their systems to be used on classified networks. This marks an important step from pilot projects toward real operational use of commercial AI in government and defense settings, giving agencies lawful, vetted access to cutting-edge capabilities.

The practical effect is immediate: analysts and operators can now leverage advanced models within secure environments, reducing friction between commercial innovation and mission needs. By enabling tested, contractor-supported AI tools on classified systems, the Defense Department can accelerate development of tailored, secure applications that enhance situational awareness, data processing, and decision support.

Security and supply-chain scrutiny played a key role in the approvals. Anthropic — previously used by the department — was left out after being designated a supply-chain risk, demonstrating that access to classified networks will follow strict vetting rather than blanket adoption. That rigorous approach builds public trust that advanced AI will be integrated responsibly and securely.

Overall, these agreements signal growing institutional confidence in commercial AI and a stronger partnership between private innovators and government. With clear guidelines and vetted providers in place, the Pentagon can more quickly field mission-ready AI tools that improve efficiency, speed intelligence workflows, and support national security priorities.

  • Who: OpenAI, Google, Microsoft, Amazon, Nvidia, xAI, Reflection.
  • What: Authorized use of AI tools in classified networks under vetted agreements.
  • Why it matters: Enables secure, practical deployment of commercial AI for defense tasks while enforcing supply-chain safeguards.

Get AI Wins in Your Inbox

The best positive AI stories delivered to your inbox. No spam, unsubscribe anytime.